Face Generation

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data

You'll be using two datasets in this project:

  • MNIST
  • CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".

In [1]:
data_dir = './data'

# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
# data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)
Found mnist Data
Found celeba Data

Explore the Data

MNIST

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.

In [2]:
show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')
Out[2]:
<matplotlib.image.AxesImage at 0x7fe7c45a0d68>

CelebA

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.

In [3]:
show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))
Out[3]:
<matplotlib.image.AxesImage at 0x7fe7d405a358>

Preprocess the Data

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network

You'll build the components necessary to build a GANs by implementing the following functions below:

  • model_inputs
  • discriminator
  • generator
  • model_loss
  • model_opt
  • train

Check the Version of TensorFlow and Access to GPU

This will check to make sure you have the correct version of TensorFlow and access to a GPU

In [4]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
TensorFlow Version: 1.8.0
Default GPU Device: /device:GPU:0

Input

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

  • Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
  • Z input placeholder with rank 2 using z_dim.
  • Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)

In [5]:
import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    # TODO: Implement Function
    inputs_real = tf.placeholder(tf.float32, (None, image_width, image_height, image_channels), name='input_real')
    inputs_z = tf.placeholder(tf.float32, (None, z_dim), name='input_z')
    learning_rate = tf.placeholder(tf.float32, name='learning_rate')
    return inputs_real, inputs_z, learning_rate


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)
Tests Passed

Discriminator

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the discriminator, tensor logits of the discriminator).

In [6]:
def discriminator(images, reuse=False):
    """
    Create the discriminator network
    :param images: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    # TODO: Implement Function
    alpha = 0.2
    with tf.variable_scope('discriminator', reuse=reuse):
        x1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same')
        relu1 = tf.maximum(alpha * x1, x1)
        
        x2 = tf.layers.conv2d(relu1, 128, 5, strides=2, padding='same')
        bn2 = tf.layers.batch_normalization(x2, training=True)
        relu2 = tf.maximum(alpha * bn2, bn2)
        
        x3 = tf.layers.conv2d(relu2, 256, 5, strides=2, padding='same')
        bn3 = tf.layers.batch_normalization(x3, training=True)
        relu3 = tf.maximum(alpha * bn3, bn3)

        # Flatten it
        flat = tf.reshape(relu3, (-1, 4*4*256))
        logits = tf.layers.dense(flat, 1)
        out = tf.sigmoid(logits)
    return out, logits


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)
Tests Passed

Generator

Implement generator to generate an image using z. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.

In [7]:
def generator(z, out_channel_dim, is_train=True):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
    # TODO: Implement Function
    alpha = 0.2
    
    # TODO not sure about reuse flag here
    with tf.variable_scope('generator', reuse=(not is_train)):
        # First fully connected layer
        x1 = tf.layers.dense(z, 7 * 7 * 512)
        # Reshape it to start the convolutional stack
        x1 = tf.reshape(x1, (-1, 7, 7, 512))
        x1 = tf.layers.batch_normalization(x1, training=is_train)
        x1 = tf.maximum(alpha * x1, x1)
        
        x2 = tf.layers.conv2d_transpose(x1, 256, 5, strides=2, padding='same')
        x2 = tf.layers.batch_normalization(x2, training=is_train)
        x2 = tf.maximum(alpha * x2, x2)
        
        x3 = tf.layers.conv2d_transpose(x2, 128, 5, strides=2, padding='same')
        x3 = tf.layers.batch_normalization(x3, training=is_train)
        x3 = tf.maximum(alpha * x3, x3)
        
        # Output layer
        logits = tf.layers.conv2d_transpose(x3, out_channel_dim, 4, strides=1, padding='same')
        out = tf.tanh(logits)
    
    return out


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)
Tests Passed

Loss

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

  • discriminator(images, reuse=False)
  • generator(z, out_channel_dim, is_train=True)
In [8]:
def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    # TODO: Implement Function
    g_model = generator(input_z, out_channel_dim)
    d_model_real, d_logits_real = discriminator(input_real)
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)
    
    smooth = 0.1
    
    d_loss_real = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_model_real) * (1- smooth)))
    d_loss_fake = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.zeros_like(d_model_fake)))
    g_loss = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.ones_like(d_model_fake)))

    d_loss = d_loss_real + d_loss_fake

    return d_loss, g_loss


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)
Tests Passed

Optimization

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).

In [9]:
def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    # TODO: Implement Function
    # Get weights and bias to update
    t_vars = tf.trainable_variables()
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    g_vars = [var for var in t_vars if var.name.startswith('generator')]

    # Optimize
    with tf.control_dependencies(tf.get_collection(tf.GraphKeys.UPDATE_OPS)):
        d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)

    return d_train_opt, g_train_opt


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)
Tests Passed

Neural Network Training

Show Output

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.

In [10]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train

Implement train to build and train the GANs. Use the following functions you implemented:

  • model_inputs(image_width, image_height, image_channels, z_dim)
  • model_loss(input_real, input_z, out_channel_dim)
  • model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.

In [11]:
def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    # TODO: Build Model
    
    img_channel_num = 3 if data_image_mode == 'RGB' else 1
    input_real, input_z, lrn_rate = model_inputs(data_shape[1], data_shape[2], img_channel_num, z_dim)
    d_loss, g_loss = model_loss(input_real, input_z, data_shape[3])
    # TODO learning rate VS lrn_rate
    d_opt, g_opt = model_opt(d_loss, g_loss, learning_rate, beta1)
    steps = 0
    show_every = 100
    
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch_i in range(epoch_count):
            for batch_images in get_batches(batch_size):
                batch_images *= 2
                # TODO: Train Model
                steps += 1
                batch_z = np.random.uniform(-1, 1, size=(batch_size, z_dim))
                
                _ = sess.run(d_opt, feed_dict={input_real: batch_images, input_z: batch_z})
                _ = sess.run(g_opt, feed_dict={input_z: batch_z, input_real: batch_images})
                
                if steps % show_every == 0:
                    # At the end of each epoch, get the losses and print them out
                    train_loss_d = d_loss.eval({input_z: batch_z, input_real: batch_images})
                    train_loss_g = g_loss.eval({input_z: batch_z})

                    print("Discriminator Loss: {:.4f}...".format(train_loss_d),
                          "Generator Loss: {:.4f}".format(train_loss_g))
                    show_generator_output(sess, 16, input_z, img_channel_num, data_image_mode)
                    

MNIST

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.

In [12]:
batch_size = 32
z_dim = 128
learning_rate = 0.0005
beta1 = 0.3


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)
Discriminator Loss: 1.2250... Generator Loss: 1.6978
Discriminator Loss: 1.1060... Generator Loss: 0.8253
Discriminator Loss: 1.0826... Generator Loss: 0.9632
Discriminator Loss: 1.2667... Generator Loss: 1.2590
Discriminator Loss: 1.1963... Generator Loss: 1.1719
Discriminator Loss: 1.3860... Generator Loss: 0.5574
Discriminator Loss: 1.3945... Generator Loss: 1.2216
Discriminator Loss: 1.0431... Generator Loss: 1.0630
Discriminator Loss: 1.9806... Generator Loss: 0.2555
Discriminator Loss: 1.6416... Generator Loss: 1.4330
Discriminator Loss: 1.1190... Generator Loss: 0.7684
Discriminator Loss: 1.4105... Generator Loss: 0.5660
Discriminator Loss: 1.4568... Generator Loss: 0.5011
Discriminator Loss: 1.0588... Generator Loss: 1.0025
Discriminator Loss: 1.2116... Generator Loss: 1.9276
Discriminator Loss: 0.9439... Generator Loss: 1.0668
Discriminator Loss: 2.5625... Generator Loss: 0.1915
Discriminator Loss: 1.2585... Generator Loss: 0.5625
Discriminator Loss: 0.9482... Generator Loss: 1.1340
Discriminator Loss: 0.8233... Generator Loss: 1.4092
Discriminator Loss: 0.7326... Generator Loss: 1.6265
Discriminator Loss: 0.8464... Generator Loss: 1.4702
Discriminator Loss: 1.6538... Generator Loss: 0.4423
Discriminator Loss: 0.9285... Generator Loss: 1.3937
Discriminator Loss: 0.9192... Generator Loss: 1.0276
Discriminator Loss: 0.7868... Generator Loss: 1.1301
Discriminator Loss: 0.8780... Generator Loss: 1.5508
Discriminator Loss: 0.8838... Generator Loss: 1.0803
Discriminator Loss: 0.5810... Generator Loss: 1.7794
Discriminator Loss: 1.5112... Generator Loss: 0.5497
Discriminator Loss: 0.7341... Generator Loss: 1.3997
Discriminator Loss: 0.7548... Generator Loss: 1.2573
Discriminator Loss: 1.4567... Generator Loss: 0.5321
Discriminator Loss: 0.7570... Generator Loss: 1.3226
Discriminator Loss: 2.1707... Generator Loss: 0.3709
Discriminator Loss: 0.7915... Generator Loss: 1.1957
Discriminator Loss: 1.7869... Generator Loss: 0.3907

CelebA

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.

In [13]:
batch_size = 32
z_dim = 128
learning_rate = 0.0005
beta1 = 0.3


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)
Discriminator Loss: 0.4076... Generator Loss: 4.4672
Discriminator Loss: 1.4985... Generator Loss: 0.4473
Discriminator Loss: 1.4892... Generator Loss: 0.4440
Discriminator Loss: 1.3359... Generator Loss: 2.1843
Discriminator Loss: 1.3957... Generator Loss: 0.6151
Discriminator Loss: 1.3095... Generator Loss: 0.5447
Discriminator Loss: 0.9534... Generator Loss: 1.6554
Discriminator Loss: 0.8373... Generator Loss: 2.2343
Discriminator Loss: 1.5560... Generator Loss: 0.4606
Discriminator Loss: 1.6483... Generator Loss: 2.5460
Discriminator Loss: 0.8026... Generator Loss: 1.0663
Discriminator Loss: 0.8422... Generator Loss: 1.2363
Discriminator Loss: 1.0263... Generator Loss: 1.6063
Discriminator Loss: 0.9906... Generator Loss: 1.4320
Discriminator Loss: 1.1079... Generator Loss: 0.9827
Discriminator Loss: 1.6818... Generator Loss: 0.3281
Discriminator Loss: 1.0268... Generator Loss: 0.7570
Discriminator Loss: 1.1949... Generator Loss: 1.0735
Discriminator Loss: 1.3504... Generator Loss: 0.5073
Discriminator Loss: 1.1469... Generator Loss: 0.9203
Discriminator Loss: 1.3109... Generator Loss: 0.8793
Discriminator Loss: 1.1239... Generator Loss: 0.8329
Discriminator Loss: 0.6588... Generator Loss: 1.9077
Discriminator Loss: 1.1669... Generator Loss: 1.1045
Discriminator Loss: 1.3811... Generator Loss: 0.8284
Discriminator Loss: 1.1095... Generator Loss: 0.9882
Discriminator Loss: 1.5224... Generator Loss: 0.4831
Discriminator Loss: 1.2930... Generator Loss: 0.9917
Discriminator Loss: 1.2008... Generator Loss: 0.8386
Discriminator Loss: 1.1303... Generator Loss: 0.9085
Discriminator Loss: 2.0016... Generator Loss: 2.4536
Discriminator Loss: 1.7961... Generator Loss: 0.3001
Discriminator Loss: 1.6109... Generator Loss: 0.3617
Discriminator Loss: 1.1554... Generator Loss: 1.4827
Discriminator Loss: 1.3195... Generator Loss: 0.7545
Discriminator Loss: 0.7975... Generator Loss: 1.1831
Discriminator Loss: 1.2915... Generator Loss: 0.5999
Discriminator Loss: 1.3936... Generator Loss: 2.1776
Discriminator Loss: 1.0983... Generator Loss: 0.9195
Discriminator Loss: 1.4748... Generator Loss: 0.4343
Discriminator Loss: 1.2607... Generator Loss: 0.9397
Discriminator Loss: 1.1177... Generator Loss: 1.3458
Discriminator Loss: 1.2972... Generator Loss: 0.7627
Discriminator Loss: 1.1083... Generator Loss: 0.9308
Discriminator Loss: 0.5710... Generator Loss: 1.9109
Discriminator Loss: 1.2270... Generator Loss: 1.3427
Discriminator Loss: 1.1805... Generator Loss: 1.0868
Discriminator Loss: 0.8439... Generator Loss: 1.2340
Discriminator Loss: 1.4934... Generator Loss: 0.4211
Discriminator Loss: 1.6627... Generator Loss: 1.7907
Discriminator Loss: 1.0377... Generator Loss: 0.8320
Discriminator Loss: 1.4668... Generator Loss: 0.4857
Discriminator Loss: 1.1052... Generator Loss: 0.7675
Discriminator Loss: 1.1471... Generator Loss: 0.6705
Discriminator Loss: 1.3601... Generator Loss: 0.5843
Discriminator Loss: 1.1656... Generator Loss: 0.8696
Discriminator Loss: 0.7885... Generator Loss: 1.2969
Discriminator Loss: 1.1523... Generator Loss: 0.8739
Discriminator Loss: 1.0919... Generator Loss: 1.1309
Discriminator Loss: 0.8820... Generator Loss: 1.1642
Discriminator Loss: 1.4812... Generator Loss: 0.5361
Discriminator Loss: 1.1222... Generator Loss: 0.7531
Discriminator Loss: 1.1812... Generator Loss: 0.6845

Submitting This Project

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.